Convexity, classification, and risk bounds

@inproceedings{Bartlett2006ConvexityCA,
  title={Convexity, classification, and risk bounds},
  author={Peter L. Bartlett and Michael I. Jordan and Jon D. McAuliffe},
  year={2006}
}
Many of the classification algorithms developed in the machine learning literature, including the support vector machine and boosting, can be viewed as minimum contrast methods that minimize a convex surrogate of the 0-1 loss function. The convexity makes these algorithms computationally efficient. The use of a surrogate, however, has statistical consequences that must be balanced against the computational virtues of convexity. To study these issues, we provide a general quantitative… CONTINUE READING

Citations

Publications citing this paper.
SHOWING 1-10 OF 611 CITATIONS

Learning Rates for Classification with Gaussian Kernels

  • Neural Computation
  • 2017
VIEW 7 EXCERPTS
CITES BACKGROUND & METHODS
HIGHLY INFLUENCED

Multiclass Classification Calibration Functions

VIEW 30 EXCERPTS
CITES BACKGROUND, METHODS & RESULTS
HIGHLY INFLUENCED

On the Consistency of AUC Pairwise Optimization

VIEW 9 EXCERPTS
CITES BACKGROUND, METHODS & RESULTS
HIGHLY INFLUENCED

Cost-sensitive Multiclass Classification Risk Bounds

VIEW 17 EXCERPTS
CITES METHODS, BACKGROUND & RESULTS
HIGHLY INFLUENCED

Cox Process Learning

VIEW 6 EXCERPTS
CITES BACKGROUND & RESULTS
HIGHLY INFLUENCED

Multicategory large-margin unified machines

  • J. Mach. Learn. Res.
  • 2013
VIEW 12 EXCERPTS
CITES BACKGROUND & METHODS
HIGHLY INFLUENCED

Statistical Consistency of Finite-dimensional Unregularized Linear Classification

VIEW 9 EXCERPTS
CITES METHODS & BACKGROUND
HIGHLY INFLUENCED

The structured elastic net for quantile regression and support vector classification

  • Statistics and Computing
  • 2012
VIEW 8 EXCERPTS
CITES BACKGROUND & METHODS
HIGHLY INFLUENCED

FILTER CITATIONS BY YEAR

2002
2019

CITATION STATISTICS

  • 119 Highly Influenced Citations

  • Averaged 55 Citations per year over the last 3 years

Similar Papers

Loading similar papers…